Mingyuan Zhou
Sr. Applied Scientist at Amazon.
Email: mingyuanzhou5@gmail.com
[LinkedIn] [Google Scholar] [CV]
Mingyuan Zhou
Sr. Applied Scientist at Amazon.
Email: mingyuanzhou5@gmail.com
[LinkedIn] [Google Scholar] [CV]
Short Bio:
I'm currently a Senior Applied Scientist at Amazon, working within EE Science to drive innovation at the intersection of science and product intelligence. My work focuses on Large Language Models (LLMs), Multimodal LLMs (M-LLMs), Vision-Language Models (VLMs), image/video generation, and foundation models for product relationships that power next-generation AI applications. Previously, I was a Senior Staff Research Scientist at InnoPeak Technology(OPPO US Research Center), focusing on the development of AIGC foundation models and advancements. I received my PhD degree in Computer Science from the University of Delaware, advised by Prof. Jingyi Yu, and my MS degree in Computer Engineering from Stevens Institute of Technology. My research interests span a range of topics in deep learning, computer vision, computer graphics, and computational photography, with a particular focus on AIGC foundation models (covering 2D image and video editing/generation, 3D generation, 4D motion generation), 3D reconstruction and rendering, non-conventional imaging, illumination and reflectance analysis, and so on.
Selected Publications:
Mingyuan Zhou, Rakib Hyder, Ziwei Xuan, Guojun Qi.
CVPR 2024
Zhangsihao Yang, Mingyuan Zhou, Mengyi Shan, Bingbing Wen, Ziwei Xuan, Mitch Hill, Junjie Bai, Guo-Jun Qi, Yalin Wang
CVPR 2024
Yuqi Ding, Yu Ji, Mingyuan Zhou, Sing Bing Kang, and Jinwei Ye.
ICCV 2021 Oral Presentation
Mingyuan Zhou, Yuqi Ding, Yu Ji, S. Susan Young, Jingyi Yu, and Jinwei Ye.
ICCP 2020, TPAMI 2020
Wei Yang, Yingliang Zhang, Jinwei Ye, Yu Ji, Zhong Li, Mingyuan Zhou, and Jingyi Yu.
TPAMI 2019
Previous Projects:
Developed a purpose-built volumetric stage, along with a comprehensive volumetric capture and 3D reconstruction solution.
Volumetric Reconstruction
Reconstructed 3D volumetric data are applied for different applications. Such data can be viewed by using mobile phones, 3D displays. AR/VR goggles, and TVs. Volumetric capture presents exciting future possibilities for mixed reality experiences.
Developed an advanced object scanning system for precise 3D object reconstruction and rendering.
Developed different human face capture systems designed to recover high-quality 3D facial geometry.